INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     barre
    -0.09
     Wilde
    -0.08
     Barre
    -0.08
     Radiation
    -0.08
    _socket
    -0.08
     vac
    -0.08
     Vac
    -0.08
     kijken
    -0.07
     dienen
    -0.07
    socket
    -0.07
    POSITIVE LOGITS
     depan
    0.09
     तेज
    0.08
    เจ้
    0.08
     मिला
    0.08
     portrait
    0.08
    587
    0.08
    एम
    0.08
     reconna
    0.08
     accommodates
    0.08
    593
    0.07
    Act Density 0.002%

    No Known Activations